Abstract
Background
Patients with invasive melanoma are at increased risk of developing subsequent invasive melanoma, but the risks for those with primary in situ melanoma are unclear.
Objectives
To assess and compare the cumulative risk of subsequent invasive melanoma after primary invasive or in situ melanoma. To estimate the standardized incidence ratio (SIR) of subsequent invasive melanoma compared to population incidence in both cohorts.
Methods
Patients with a first diagnosis of melanoma (invasive or in situ) between 2001 and 2017 were identified from the New Zealand national cancer registry, and any subsequent invasive melanoma during follow‐up to the end of 2017 identified. Cumulative risk of subsequent invasive melanoma was estimated by Kaplan–Meier analysis separately for primary invasive and in situ cohorts. Risk of subsequent invasive melanoma was assessed using Cox proportional hazard models. SIR was assessed, allowing for age, sex, ethnicity, year of diagnosis and follow up time.
Results
Among 33 284 primary invasive and 27 978 primary in situ melanoma patients, median follow up time was 5.5 and 5.7 years, respectively. A subsequent invasive melanoma developed in 1777 (5%) of the invasive and 1469 (5%) of the in situ cohort, with the same median interval (2.5 years) from initial to first subsequent lesion in both cohorts. The cumulative incidence of subsequent invasive melanoma at 5 years was similar in the two cohorts (invasive 4.2%, in situ 3.8%); the cumulative incidence increased linearly over time in both cohorts. The risk of subsequent invasive melanoma was marginally higher for primary invasive compared to in situ melanoma after adjustment for age, sex, ethnicity and body site of the initial lesion (hazard ratio 1.11, 95% CI 1.02–1.21). Compared to population incidence, the SIR of invasive melanoma was 4.6 (95% CI 4.3–4.9) for the primary invasive and 4 (95% CI 3.7–4.2) for the primary in situ melanoma cohorts.
Conclusions
The risk of subsequent invasive melanoma is similar whether patients present with in situ or invasive melanoma. Thus follow‐up surveillance for new lesions should be similar, although patients with invasive melanoma require more surveillance for recurrence.
1.
What is already known about this topic?
Patients with primary invasive melanoma are at increased risk of developing subsequent invasive melanoma but it is unclear how this risk compares with that for patients with primary in situ melanoma.
What does this study add?
The cumulative risk of subsequent invasive melanoma at 5 years' follow‐up was similar among primary invasive (4.2%) and in situ (3.8%) melanoma patients.
The standardized incidence ratio (SIR) of subsequent invasive melanoma was increased and of similar magnitude, with SIRs of 4.6 (95% CI, 4.3–4.9) for primary invasive and 4 (95% CI, 3.7–4.2) for in situ melanoma patients compared to population incidence.
Patients with primary in situ (as well as invasive) melanoma should be considered for follow‐up surveillance for new lesions.
2. INTRODUCTION
Melanoma is the 19th most common cancer globally 1 and the third most common cancer in New Zealand. 2 The incidence of melanoma has been increasing worldwide over the past 50 years, especially in fair‐skinned populations and people living in areas of lower latitude. 3 New Zealand and Australia experienced the greatest burden of melanoma followed by North America and European countries in 2013 4 and 2015. 5
Melanoma is diagnosed as either an in situ or invasive lesion. It has been suggested that simple excision of in situ lesions might prevent progression to invasive lesions and metastasis. 6 The annual incidence of in situ melanoma has increased at a faster rate to that of invasive melanoma in Australia,7, 8, 9, 10 New Zealand,11, 12 Italy,13, 14, 15 Spain 16 and the United States.17, 18, 19 However, the relationship between in situ and invasive melanoma remains unclear 11 even though many previous studies have compared their epidemiological characteristics7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29 or subsequent risk of melanoma.27, 30, 31, 32, 33, 34 The risk of subsequent melanoma of either type was found to be higher among the patients with primary in situ melanoma when compared to those with primary invasive melanoma for some studies,30, 31, 32, 33, 34 whereas a Queensland study 27 found that the risk of subsequent invasive melanoma was greater for patients with primary invasive melanoma compared to those with primary in situ melanoma. The aim of this study was to determine whether the incidence of subsequent invasive melanoma differed according to primary melanoma type (in situ or invasive), and whether the incidence varied by key characteristics (sex, age groups, ethnicity and body site of primary melanoma), using national population‐based cohorts.
3. PATIENTS AND METHODS
Patients with their first diagnosed melanoma (either in situ or invasive) diagnosed from 1 January 2001 to 31 December 2017 were identified from the New Zealand Cancer Registry (NZCR). NZCR 35 is a population‐based, legislatively mandated, cancer registry in New Zealand which receives all pathology reports of malignant tumours. In situ melanoma pathology reports are received and coded in the same way as invasive reports, although data on in situ lesions are not routinely reported. Data were linked to subsequent melanoma diagnoses (also from the NZCR) and/or death (from national mortality data collection) during follow‐up (up to 31 December 2017), and de‐identified data then used for analysis. Follow up time was from 3 months after the initial diagnosis date, as patients with synchronous diagnosis within 90 days were excluded for follow‐up analysis. A few previous studies of follow up made a similar adjustment.33, 36 The restriction was applied because synchronous diagnosis may have been influenced by the health care attention the patient received, or may be related to the primary lesion. An event was categorized as ‘1’ for subsequent diagnosis of invasive melanoma after primary melanoma (either invasive or in situ) whereas censoring was categorized as ‘0’ at the time of death or at the end of study period. Thirty‐four patients whose sex and ethnicity were unknown, or whose domicile code was invalid were excluded from the study.
The cumulative incidence of subsequent primary invasive melanoma after diagnosis with primary in situ or invasive melanoma was assessed by non‐parametric Kaplan–Meier analysis at 5, 10 and 15 years of follow‐up time by key characteristics such as sex, age group, ethnicity and body site, separately by primary melanoma type (invasive or in situ).
The cumulative risk differences were tested by the log‐rank test within key characteristics of each primary melanoma type and between the two cohorts of primary melanoma for each key characteristic. The risks of subsequent invasive melanoma among the primary invasive and in situ melanoma cohorts were compared according to key characteristics (univariate and multivariate) in Cox proportional hazard models. Univariate models assessed the impact of covariates, and multivariate model measured the impact of other covariates when the risk was estimated for the one factor after adjustment. 37 The proportionality assumption in each model was assessed by using the global Schoenfeld test 38 and plotting log[−log(survival)] versus log(time).
The incidence of subsequent invasive melanoma relative to population incidence was estimated by standardized incidence ratio (SIR). This analysis was done, separately for each sex, for patients aged between 30 and 74 years in the two largest groups in New Zealand statistics, the ‘Other’ (predominantly New Zealand European) ethnic group and the Māori ethnic group. A more limited age range was used because melanoma in younger and older subjects may differ both in biology and diagnosis. The observed number of subsequent invasive melanoma from each cohort of primary melanoma type was compared with the expected number of subsequent invasive melanoma, taking into account year of diagnosis, exact follow up duration, sex, age, ethnicity and body site of primary melanoma. A detailed explanation of the calculation is provided in Appendix S1 and Table S1.
4. RESULTS
This population‐based cohort study included 61 262 eligible patients with a first diagnosed invasive or in situ melanoma from 1 January 2001 to 31 December 2017; 33 284 (54%) patients made up the cohort with an initial primary invasive melanoma and 27 978 (46%) patients formed the cohort with initial primary in situ melanoma. Median follow up time was 5.5 years (interquartile range, 2.1–10 years) for invasive cohort and 5.7 years (interquartile range, 2.4–9.9 years) for in situ cohort with the same interval of 2.5 median years between initial and subsequent diagnoses. The same proportion (5%) of subsequent invasive melanoma developed with 1777 patients in primary invasive melanoma and 1469 patients in primary in situ melanoma, excluding the synchronous diagnosis of melanoma within 90 days. There were 1075 cases of subsequent invasive cases (550 cases from invasive cohort, 525 cases from in situ cohort) developed within 90 days of their first melanomas.
Table 1 describes the key characteristics of patients in the primary invasive and in situ melanoma cohorts. Patients with invasive melanoma were slightly younger at diagnosis than those with in situ melanoma (median age 63 vs. 64 years, p < 0.001 by the Wilcoxon–Mann–Whitney test), though this difference is unlikely to be clinically significant. Majority of primary melanoma developed in 60–69 years age groups and few lesions were found in extreme age groups. Males had a higher proportion of both types of primary melanoma than females. Most melanomas were found in European/Other ethnic group. Trunk was the most common region for both types of primary melanomas followed by upper and lower limbs in primary invasive, and face and upper limbs in primary in situ cohort.
TABLE 1.
Key characteristics of patients diagnosed with either primary invasive or in situ melanoma
| Key characteristics | Primary invasive melanoma, n (%) | Primary in situ melanoma, n (%) |
|---|---|---|
| 33 284 | 27 978 | |
| Age at diagnosis | ||
| Mean (SD) | 62.2 (16) | 63.1 (14.3) |
| Median (IQR) | 63 (51–74) | 64 (54–74) |
| Age groups | ||
| 10–19 | 121 (0%) | 58 (0%) |
| 20–29 | 759 (2%) | 390 (1%) |
| 30–39 | 2117 (6%) | 1243 (4%) |
| 40–49 | 4324 (13%) | 3181 (11%) |
| 50–59 | 6551 (20%) | 5773 (21%) |
| 60–69 | 7608 (23%) | 7504 (27%) |
| 70–79 | 6734 (20%) | 6212 (22%) |
| 80–89 | 4234 (13%) | 3194 (11%) |
| 90+ | 834 (3%) | 423 (2%) |
| Sex | ||
| Female | 15 985 (48%) | 13 563 (49%) |
| Male | 17 299 (52%) | 14 415 (51%) |
| Ethnicity | ||
| Māori | 633 (2%) | 447 (2%) |
| Pacific | 104 (0%) | 48 (0%) |
| Asian | 64 (0%) | 39 (0%) |
| European/Other | 32 483 (98%) | 27 444 (98%) |
| Body sites of first melanoma | ||
| Face | 3353 (10%) | 6962 (25%) |
| Scalp and neck | 2034 (6%) | 1721 (6%) |
| Trunk | 10 183 (31%) | 7952 (28%) |
| Upper limbs | 7576 (23%) | 6345 (23%) |
| Lower limbs | 8605 (26%) | 4855 (17%) |
| Other area | 1525 (5%) | 123 (0%) |
| Overlapping area | 8 (0%) | 20 (0%) |
4.1. Cumulative incidence of subsequent invasive melanoma
The overall cumulative risks of subsequent invasive melanoma were 4% at 5 years, 8% at 10 years and 11.7% at 15 years of follow‐up time. Table 2 describes and compares the cumulative incidence of subsequent invasive melanoma between a cohort of patients with first invasive and a cohort of patients with in situ melanoma, and between each key patient characteristic (age, sex, ethnicity and body site) for each cohort. The risks between the two cohorts were closely similar with 4.2% and 8.2% for invasive cohort and 3.8% and 7.7% for in situ cohort at 5 and 10 years of follow‐up. No statistical difference in risks was noted between the two cohorts (p = 0.508). The risks were significantly different within sex, ethnic groups, age groups and body sites of primary melanoma for each cohort. The risks were higher in males, in the NZ European population, in older age groups and in the scalp and neck region; the associations with these factors were similar in the two cohorts.
TABLE 2.
The cumulative risk of subsequent invasive melanoma at 5, 10 and 15 years of follow‐up and a long‐rank test with p value
| Key characteristics | Number of subsequent invasive melanoma | Cumulative risk (%) | p‐Value | ||
|---|---|---|---|---|---|
| 5‐year | 10‐year | 15‐year | |||
| Primary melanoma type | |||||
| Invasive | 1469 | 4.2 | 8.2 | 11.7 | 0.508 |
| In situ | 1777 | 3.8 | 7.7 | 11.7 | |
| Primary invasive melanoma | |||||
| Sex | |||||
| Female | 699 | 3.3 | 6.4 | 8.8 | 0.000 |
| Male | 1078 | 5.1 | 10.0 | 15.0 | |
| Ethnicity | |||||
| European/Other | 1739 | 4.3 | 8.3 | 12.0 | 0.004 |
| Māori | 17 | 2.2 | 5.0 | 5.7 | |
| Age groups | |||||
| 20–29 | 12 | 0.8 | 1.7 | 2.5 | 0.000 |
| 30–39 | 67 | 1.7 | 3.6 | 5.6 | |
| 40–49 | 161 | 2.5 | 4.7 | 6.5 | |
| 50–59 | 355 | 3.6 | 7.2 | 11.5 | |
| 60–69 | 456 | 4.3 | 9.2 | 14.2 | |
| 70–79 | 493 | 6.4 | 12.8 | 18.6 | |
| 80–89 | 203 | 5.9 | 11.8 | 15.8 | |
| Body sites of primary melanoma | |||||
| Face | 200 | 5.2 | 9.9 | 12.7 | 0.000 |
| Scalp and neck | 620 | 5.1 | 10.5 | 14.9 | |
| Trunk | 386 | 4.7 | 9.0 | 12.9 | |
| Upper limbs | 408 | 3.6 | 7.8 | 12.0 | |
| Lower limbs | 126 | 3.6 | 6.5 | 9.5 | |
| Primary in situ melanoma | |||||
| Sex | |||||
| Female | 596 | 3.0 | 6.1 | 9.3 | 0.000 |
| Male | 873 | 4.6 | 9.3 | 14.3 | |
| Ethnicity | |||||
| European/Other | 1450 | 3.9 | 7.9 | 12.0 | 0.006 |
| Māori | 11 | 2.5 | 3.6 | 3.6 | |
| Age groups | |||||
| 20–29 | 5 | 0.9 | 1.8 | 1.8 | 0.000 |
| 30–39 | 31 | 1.9 | 2.7 | 4.9 | |
| 40–49 | 95 | 1.5 | 3.9 | 5.9 | |
| 50–59 | 247 | 2.4 | 5.8 | 8.9 | |
| 60–69 | 447 | 4.3 | 8.3 | 14.4 | |
| 70–79 | 436 | 5.2 | 11.7 | 17.7 | |
| 80–89 | 191 | 6.5 | 12.4 | 16.5 | |
| Body sites of first melanoma | |||||
| Face | 421 | 4.3 | 8.5 | 12.6 | 0.000 |
| Scalp and neck | 390 | 6.0 | 11.0 | 16.7 | |
| Trunk | 316 | 3.5 | 7.8 | 12.6 | |
| Upper limbs | 216 | 3.6 | 7.5 | 11.2 | |
| Lower limbs | 119 | 3.2 | 6.0 | 8.8 | |
The cumulative incidence of subsequent invasive melanoma was found to increase constantly, with no evidence of a decrease or an increase over time. The risks, and the patterns of risk for the two cohorts were very similar (Figure 1a). This constant linear increase in risks was noted in both sexes (Figure 1b), European population (Figure 1c), in trunk, upper limbs and lower limbs of primary melanoma sites (Figure 1d,e) and patients aged between 50 and 79 years (Figure 1f,g). The risk patterns were found to be similar between invasive and in situ melanoma across sexes, ethnic groups, body sites and age groups except in face, scalp and neck and in extreme of ages (less than 30 and more the 80 years; Figure 1a–g).
FIGURE 1.
Cumulative risk of subsequent invasive melanoma at 5, 10 and 15 years of follow‐up since after their first diagnosis of primary melanoma (a) by primary melanoma type (b) by sex and primary melanoma type (c) by ethnic group and primary melanoma type (d) by body sites of primary invasive melanoma (e) by body sites of primary in situ melanoma (f) by age groups of invasive melanoma (g) by age groups of in situ melanoma
4.2. Comparison between invasive and in situ cohorts
The risk difference for subsequent invasive melanoma was compared between primary invasive and in situ melanoma cohorts (in situ as a reference) and for male, female, European, Māori, each body site and each age group in terms of univariate analysis (Table 3). The subsequent risk of invasive melanoma for patients with primary invasive melanoma was 7% higher than those with primary in situ melanoma in an unadjusted analysis but the difference was not statistically significant (95% CI 0.98–1.16, Table 3). No differences in unadjusted hazard risk between invasive and in situ cohorts were observed in more detailed analysis by sub‐categories except in the 50–59‐year‐old age group; the risk was 27% higher for invasive melanoma than in situ lesions among this age group.
TABLE 3.
Risk difference for subsequent invasive melanoma between primary invasive and in situ melanoma cohort (in situ as a reference) by Log‐rank test and the unadjusted hazard ratios (HRs) for the cohort of primary invasive melanoma compared to the cohort of in situ melanoma by each sub‐category of key characteristics
| Key characteristics | Log‐rank test (p value) for differences between primary invasive and in situ melanoma | Unadjusted HRs (95% CIs) for primary invasive compared to in situ melanoma |
|---|---|---|
| Primary melanoma type | 0.118 | 1.07 (0.98–1.16) |
| Sex | ||
| Female | 0.170 | 1.10 (0.96–1.25) |
| Male | 0.382 | 1.05 (0.94–1.17) |
| Ethnicity | ||
| European/Other | 0.144 | 1.06 (0.98–1.16) |
| Māori | 0.138 | 2.17 (0.78–6.02) |
| Body sites of first melanoma | ||
| Face | 0.092 | 1.22 (0.97–1.54) |
| Scalp and neck | 0.979 | 1.00 (0.74–1.36) |
| Trunk | 0.103 | 1.13 (0.98–1.30) |
| Upper limbs | 0.808 | 0.98 (0.82–1.16) |
| Lower limbs | 0.338 | 1.10 (0.91–1.32) |
| Age groups | ||
| 30–39 | 0.367 | 1.22 (0.79–1.87) |
| 40–49 | 0.225 | 1.17 (0.91–1.51) |
| 50–59 | 0.005 | 1.27 (1.08–1.49) |
| 60–69 | 0.812 | 1.02 (0.89–1.16) |
| 70–79 | 0.095 | 1.17 (0.97–1.40) |
4.3. Factors related to risks of melanoma—multivariate analysis
The adjusted hazard ratios (HRs) for subsequent invasive melanoma were compared between patients with a first invasive and a first in situ melanoma (Table 4). The risk of subsequent invasive melanoma was 11% (95% CI 1.02–1.21) higher in primary invasive cohort compared to in situ cohort after adjustment for age, sex, ethnicity and body site. The risk increased by 4% with each increase in age at diagnosis (in years), and males had a 38% higher risk than females for total primary melanoma and also for the two cohorts separately. Compared to the European population, Māori experienced 47% lower risk of subsequent invasive melanoma among total primary melanoma and the risk was reduced by 67% in the primary in situ cohort. Scalp and neck was noted as the body site with highest risk for subsequent invasive melanoma compared to face for total primary melanoma cohort and the two cohorts.
TABLE 4.
Adjusted hazard ratios (HRs) with 95% CI for subsequent invasive melanoma between patients with primary invasive and in situ melanoma
| Key characteristics | Total primary melanoma cohort (in situ + invasive) | Primary in situ melanoma | Primary invasive melanoma |
|---|---|---|---|
| Primary melanoma type | |||
| Primary in situ melanoma | 1 | ‐ | ‐ |
| Primary invasive melanoma | 1.11 (1.02–1.21) | ‐ | ‐ |
| Age, per year | 1.04 (1.03–1.04) | 1.04 (1.03–1.05) | 1.03 (1.03–1.04) |
| Sex | |||
| Female | 1 | 1 | 1 |
| Male | 1.38 (1.27–1.52) | 1.43 (1.26–1.63) | 1.34 (1.19–1.51) |
| Ethnicity | |||
| European/Other | 1 | 1 | 1 |
| Māori | 0.53 (0.34–0.84) | 0.33 (0.14–0.80) | 0.69 (0.40–1.16) |
| Body sites of first melanoma | |||
| Face | 1 | 1 | 1 |
| Scalp and neck | 1.50 (1.24–1.81) | 1.58 (1.23–2.04) | 1.35 (1.02–1.79) |
| Trunk | 1.19 (1.04–1.36) | 1.22 (1.03–1.45) | 1.10 (0.89–1.37) |
| Upper limbs | 1.13 (0.98–1.30) | 1.26 (1.05–1.51) | 0.98 (0.78–1.23) |
| Lower limbs | 1.07 (0.92–1.24) | 1.18 (0.96–1.45) | 0.94 (0.75–1.18) |
Note: Column 3 compares the risk of subsequent invasive melanoma by each key patient characteristic (age, sex, ethnicity and body site) after adjustment for covariates among patients with primary in situ melanoma. This is the same for column 4, for primary invasive melanoma cohort.
4.4. Risk of subsequent invasive melanoma relative to population incidence
The risk of subsequent invasive melanoma was estimated by SIR (ratio of observed cases to expected cases in the general population) for those aged between 30 and 74 years, restricted to the Māori and European population (Table 5). The subsequent invasive melanomas were more likely to develop in primary invasive cohort compared to the primary in situ cohort, which was statistically significant (incidence rate ratio (IRR) 1.2, 95% CI 1.1–1.3). This pattern was similar in both sexes, Māori and European population, those aged 50–59 years, and in face and upper limbs of primary melanoma in further detailed analysis by sub‐categories (Table S2).
TABLE 5.
SIRs and IRRs of subsequent invasive melanoma among primary invasive and in situ melanoma cohort by sex and ethnicity, and by age groups
| Key characteristics | Observed (O) | Expected (E) | Standardized incidence ratio, SIR (95% CIs) | Incidence rate ratio, IRR (95% CIs) |
|---|---|---|---|---|
| Primary melanoma type | ||||
| Primary invasive | 1273 | 277.2 | 4.6 (4.3–4.9) | 1.2 (1.1–1.3) |
| Primary in situ | 1032 | 261.2 | 4.0 (3.7–4.2) | |
| Primary invasive melanoma | ||||
| Sex | ||||
| Male | 751 | 168.3 | 4.5 (4.1–4.8) | 0.9 (0.8–1.0) |
| Female | 522 | 108.9 | 4.8 (4.4–5.2) | |
| Ethnicity | ||||
| Māori | 14 | 1.4 | 9.9 (5.4–16.7) | 2.2 (1.0–3.3) |
| European/Other | 1259 | 275.8 | 4.6 (4.3–4.8) | |
| Age groups | ||||
| 30–39 | 67 | 8.7 | 7.7 (6.0–9.8) | 1.9 (1.4–2.4) |
| 40–49 | 157 | 27.4 | 5.7 (4.9–6.7) | 1.4 (1.2–1.7) |
| 50–59 | 352 | 62.0 | 5.7 (5.1–6.3) | 1.4 (1.2–1.6) |
| 60–69 | 442 | 108.2 | 4.1 (3.7–4.5) | Reference |
| 70–79 | 255 | 71.0 | 3.6 (3.2–4.1) | 0.9 (0.8–1.0) |
| Body sites of first melanoma | ||||
| Face | 105 | 25.1 | 4.2 (3.4–5.1) | Reference |
| Scalp and neck | 315 | 72.4 | 4.4 (3.9–4.9) | 1.0 (0.8–1.3) |
| Trunk | 88 | 15.2 | 5.8 (4.6–7.1) | 1.4 (1.0–1.8) |
| Upper limbs | 477 | 99.3 | 4.8 (4.4–5.3) | 1.1 (0.9–1.4) |
| Lower limbs | 288 | 65.2 | 4.4 (3.9–5.0) | 1.1 (0.9–1.3) |
| Primary in situ melanoma | ||||
| Sex | ||||
| Male | 616 | 161.1 | 3.8 (3.5–4.1) | 0.9 (0.8–1.0) |
| Female | 416 | 100.1 | 4.2 (3.8–4.6) | |
| Ethnicity | ||||
| Māori | 5 | 1.3 | 3.9 (1.2–9.3) | 1.0 (0.1–1.9) |
| European/Other | 1027 | 260.0 | 4.0 (3.7–4.2) | |
| Age groups | ||||
| 30–39 | 31 | 5.3 | 5.8 (4.0–8.3) | 1.5 (0.9–2.0) |
| 40–49 | 94 | 19.5 | 4.8 (3.9–5.9) | 1.2 (0.9–1.5) |
| 50–59 | 245 | 54.9 | 4.5 (3.9–5.1) | 1.1 (0.9–1.3) |
| 60–69 | 446 | 111.3 | 4.0 (3.6–4.4) | Reference |
| 70–79 | 216 | 70.1 | 3.1 (2.7–3.5) | 0.8 (0.7–0.9) |
| Body sites of first melanoma | ||||
| Face | 233 | 72.1 | 3.2 (2.8–3.7) | Reference |
| Scalp and neck | 169 | 40.9 | 4.1 (3.5–4.8) | 1.3 (1.0–1.6) |
| Trunk | 82 | 16.0 | 5.1 (4.1–6.4) | 1.6 (1.2–2.0) |
| Upper limbs | 303 | 74.5 | 4.1 (3.6–4.6) | 1.3 (1.1–1.5) |
| Lower limbs | 245 | 57.7 | 4.2 (3.7–4.8) | 1.3 (1.1–1.6) |
The SIRs in male were 4.5 for invasive and 3.8 for in situ cohort while it was 4.8 in invasive and 4.2 in in situ melanomas of females (Table 5). The IRR in males was 10% lower in both cohorts compared to females which was not statistically significant (95% CI 0.8–1). The IRR of Māori was 120% higher significantly than that of European for invasive melanoma, but for in situ cohort there was no significant difference between ethnic groups (ratio 1, 95% CI 0.1–1.9). Subsequent invasive melanoma were more likely to appear in the younger age groups in both types of primary melanoma. The SIRs of primary melanoma in trunk was 5.8 for invasive and 5.1 for in situ melanomas. Therefore, patients with primary melanoma in trunk were more likely to be diagnosed with subsequent invasive melanoma than those with primary lesion on face for invasive cohort, but for in situ cohort, the IRR of primary lesions at any regions of the body was higher than that on face.
5. DISCUSSION
This study found that the cumulative incidence of subsequent invasive melanoma increased similarly for both primary invasive and in situ melanoma cohorts. The risk of subsequent invasive melanoma was marginally higher for primary invasive compared to in situ melanoma after adjustment for age, sex, ethnicity and body site of the initial lesion. While we found little difference in risk of subsequent invasive melanoma according to whether primary melanoma was invasive or in situ, we did find differences in the risk of this outcome according to other factors: sex, ethnicity, age groups and body sites of first melanoma. Risk was higher in men than women, in Europeans than Māori and in scalp and neck than face among primary lesion.
The 5‐year cumulative risk of subsequent melanoma observed in our study conducted in New Zealand, 2001–2017 (4.2% invasive, 3.8% in situ) is similar to that observed in a previous study conducted in Australia (3.8‐5.0% invasive, 3.6‐6.7% in situ), 30 but higher than the findings of studies conducted in the Netherlands (1.8% invasive, 2.1% in situ) 31 and the United States (1.2% invasive, 1.3% in situ). 32 Likewise, the SIR of subsequent invasive melanoma in our study (4.6% invasive, 4.0% in situ) was similar to previous studies27, 30 in Queensland (SIRs = 5.4–8.2 invasive, 4.6–7.4 in situ). In the Dutch study, 31 the SIRs (12.4 invasive, 15.4 in situ) was much higher than that identified in Australian studies. This much higher SIRs of subsequent invasive melanoma compared with the general population and the lower risk of cumulative incidence observed in Netherlands compared with our study and those in Queensland may be because the Netherlands22, 39 has lower background incidence rates of melanoma than in Australia7, 8, 9, 10 and in New Zealand.11, 12
Multiple factors influence the identification of subsequent invasive melanoma following primary melanoma including follow‐up surveillance. 40 The aims of follow‐up have included identification of recurrence or new melanomas41, 42, 43 and also to provide psychosocial support and training for self‐examination. 43 Follow‐up guidelines and schedules are not standardized and vary considerably between countries.41, 43, 44 A previous systematic review found that schedules varied according to country, physician speciality and stage of disease. 44
Some guidelines do not mention42, 45 or do not include 46 in situ melanoma as part of their advice for melanoma follow‐up, despite the fact that, as with our study, previously published studies27, 30, 31, 32, 34 have found that the risk of subsequent invasive melanoma is similar whether the primary lesion is in situ or invasive. In New Zealand, there are no differences in follow‐up surveillance after primary invasive or in situ melanoma. 43 Follow‐up schedules depend on patient risk and needs, especially disease staging. Patients are trained to undertake self‐examination, but the responsibility for melanoma follow‐up remains with a lead health care professional (specialist, general practitioner, nurse practitioner or a combination) experienced in melanoma diagnosis and management. 43
A recent study suggested follow‐up surveillance should be based on individual risk assessment in order to adequately balance potential benefits and costs (direct surveillance costs and patient harms). 40 At the same time, results from a pilot of the MELanoma SELF surveillance (MEL‐SELF) randomized clinical trial47, 48 indicate that patient‐led surveillance for localized melanoma appears likely to be a safe, feasible and cost‐effective alternative to clinician‐led surveillance. Full results from the main MEL‐SELF trial will be available following completion of the trial in 2023.
This is the nation‐wide population‐based study that compared the subsequent risk of invasive melanoma between primary invasive and in situ melanoma patients. This study has considered key demographic characteristics (age, sex and ethnicity) as well as body site in estimating absolute and SIR of subsequent invasive melanoma among primary melanoma patients. One limitation of our ability to investigate subsequent invasive melanoma risk is that we were unable to assess the effect of other relevant factors (especially histopathological, phenotypical and genomic) as these data are not yet routinely included in the National Cancer Registry. This study was limited to the patients who are New Zealand residents and a limitation would be the failure to include international emigrants who were diagnosed with their first melanoma (either invasive or in situ) in New Zealand but had their subsequent melanoma diagnosed in other countries. However, in the relevant age ranges, permanent emigration from New Zealand is uncommon. 49 In this study we focussed only on patients with their first primary melanoma (either invasive or in situ) who were diagnosed with a first subsequent invasive melanoma and all analyses were considered at patient‐level. The consideration of further subsequent lesions can be influenced by differences in follow‐up practices and in recording.
In conclusion, our study suggests that the risk of a subsequent invasive melanoma are similar for patients presenting with either a primary invasive or in situ melanoma, since the absolute risk and SIRs of subsequent invasive melanoma were similar in both cohorts. Thus follow‐up surveillance for new lesions should be similar, although patients with invasive melanoma require more frequent surveillance for recurrence.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
ETHICS APPROVAL
This study is approved by the University of Auckland Human Participants Ethics Committee (UAHPEC) with Ref. 024010.
AUTHOR CONTRIBUTIONS
Thu Thu Win Myint: Data curation; Formal analysis; Methodology; Software; Validation; Visualization; Writing – original draft; Writing – review & editing. Vanessa Selak: Methodology; Software; Supervision; Validation; Writing – review & editing. Mark Elwood: Conceptualization; Methodology; Supervision; Validation; Writing – review & editing.
Supporting information
Supporting Information S1
ACKNOWLEDGEMENT
We thank the NZ Cancer Registry (Ministry of Health) for providing cancer registry data.
Win Myint TT, Selak V, Elwood M. The risk of subsequent invasive melanoma after a primary in situ or invasive melanoma in a high incidence country (New Zealand). Skin Health Dis. 2023;3(2):e116. 10.1002/ski2.116
DATA AVAILABILITY STATEMENT
The data for this study were provided by the New Zealand Ministry of Health, and may be available to other researchers who meet data access requirements. Please contact data_enquiries@moh.govt.nz for further details on eligibility and data provision.
References
REFERENCES
- 1. World Cancer Research Fund International . Skin cancer statistics. Available from: https://www.wcrf.org/dietandcancer/cancer‐trends/skin‐cancer‐statistics [Google Scholar]
- 2. Ministry of Health . New cancer registrations 2018. 2020. [cited 2020 Dec 29]. Available from: https://www.health.govt.nz/publication/new‐cancer‐registrations‐2018 [Google Scholar]
- 3. Matthews NH, Li WQ, Qureshi AA, Weinstock MA, Cho E. Epidemiology of melanoma. In: Ward WH, Farma JM, editors. Cutaneous melanoma: etiology and therapy. Brisbane: Codon Publications; 2017. p. 3–22. [PubMed] [Google Scholar]
- 4.Karimkhani C, Dellavalle RP, Coffeng LE, Flohr C, Hay RJ, Langan SM, et al. Global skin disease morbidity and mortality: An update from the global burden of disease study 2013. JAMA Dermatol. 2017;153:406–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Karimkhani C, Green AC, Nijsten T, Weinstock MA, Dellavalle RP, Naghavi M, et al. The global burden of melanoma: results from the Global Burden of Disease Study 2015. Br J Dermatol. 2017;177:134–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Ackerman AB. Malignant melanoma in situ: the flat, curable stage of malignant melanoma. Pathology. 1985;17:298–300. [DOI] [PubMed] [Google Scholar]
- 7. Coory M, Baade P, Aitken J, Smithers M, McLeod GRC, Ring I. Trends for in situ and invasive melanoma in Queensland, Australia, 1982–2002. Cancer Causes Control. 2006;17:21–7. [DOI] [PubMed] [Google Scholar]
- 8. Aitken JF, Youlden DR, Baade PD, Soyer HP, Green AC, Smithers BM. Generational shift in melanoma incidence and mortality in Queensland, Australia, 1995–2014. Int J Cancer. 2018;142:1528–35. [DOI] [PubMed] [Google Scholar]
- 9. Buettner PG, MacLennan R. Geographical variation of incidence of cutaneous melanoma in Queensland. Aust J Rural Health. 2008;16:269–77. [DOI] [PubMed] [Google Scholar]
- 10. Curchin DJ, Forward E, Dickison P, Harris VR, McCormack CJ, Smith SD. The acceleration of melanoma in situ: a population‐based study of melanoma incidence trends from Victoria, Australia, 1985‐2015. J Am Acad Dermatol. 2019;80:1791–3. [DOI] [PubMed] [Google Scholar]
- 11. Elwood M, Kim SJH, Ip KHK, Oakley A, Rademaker M. In situ and invasive melanoma in a high‐risk, New Zealand, population: a population‐based study. Australas J Dermatol. 2019;60:38–44. [DOI] [PubMed] [Google Scholar]
- 12. Salmon PJM, Chan WC, Griffin J, McKenzie R, Rademaker M. Extremely high levels of melanoma in Tauranga, New Zealand: possible causes and comparisons with Australia and the northern hemisphere. Australas J Dermatol. 2007;48:208–16. [DOI] [PubMed] [Google Scholar]
- 13. Ambrosini‐Spaltro A, Dal Cappello T, Deluca J, Carriere C, Mazzoleni G, Eisendle K. Melanoma incidence and Breslow tumour thickness development in the central Alpine region of South Tyrol from 1998 to 2012: a population‐based study. J Eur Acad Dermatol Venereol. 2015;29:243–8. [DOI] [PubMed] [Google Scholar]
- 14. Crocetti E, Caldarella A, Chiarugi A, Nardini P, Zappa M. The thickness of melanomas has decreased in central Italy, but only for thin melanomas, while thick melanomas are as thick as in the past. Melanoma Res. 2010;20:422–6. [DOI] [PubMed] [Google Scholar]
- 15. Pellacani G, Lo Scocco G, Vinceti M, Albertini G, Raccagni AA, Baldassari L, et al. Melanoma epidemic across the millennium: time trends of cutaneous melanoma in Emilia‐Romagna (Italy) from 1997 to 2004. J Eur Acad Dermatol VenereolV. 2008;22:213–8. [DOI] [PubMed] [Google Scholar]
- 16. Rubio‐Casadevall J, Puig‐Vives M, Puigdemont M, Vilardell L, Carbo-Bague A, Marcos‑Gragera R, et al. Patterns of increased incidence and survival of cutaneous melanoma in Girona (Spain) 1994‐2013: a population‐based study. Clin Transl Oncol. 2018;20:1617–25. [DOI] [PubMed] [Google Scholar]
- 17. Cockburn MG, Zadnick J, Deapen D. Developing epidemic of melanoma in the Hispanic population of California. Cancer. 2006;106:1162–8. [DOI] [PubMed] [Google Scholar]
- 18. Lee JA. The systematic relationship between melanomas diagnosed in situ and when invasive. Melanoma Res. 2001;11:523–9. [DOI] [PubMed] [Google Scholar]
- 19. Wei EX, Qureshi AA, Han J, Li TY, Cho E, Lin JY, et al. Trends in the diagnosis and clinical features of melanoma in situ (MIS) in US men and women: a prospective, observational study. J Am Acad Dermatol. 2016;75:698–705. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Crocetti E, Guzzinati S, Paci E, Falcini F, Zanetti R, Vercelli M, et al. Strong seasonality in the diagnosis of skin melanoma in Italy: the Italian Network of Cancer Registries (AIRTUM) study. Tumori. 2009;95:665–8. [DOI] [PubMed] [Google Scholar]
- 21. Barbe C, Hibon E, Vitry F, Le Clainche A, Grange F. Clinical and pathological characteristics of melanoma: a population‐based study in a French regional population. J Eur Acad Dermatol Venereol. 2012;26:159–64. [DOI] [PubMed] [Google Scholar]
- 22. Sacchetto L, Zanetti R, Comber H, Bouchardy C, Brewster DH, Broganelli P, et al. Trends in incidence of thick, thin and in situ melanoma in Europe. Eur J Cancer. 2018;92:108–18. [DOI] [PubMed] [Google Scholar]
- 23. Paul J, Sbicca JA, Hirano SA, Mushti S, Pariser RJ. Changing age distribution of melanoma patients: a 22‐year, single‐site perspective. South Med J. 2013;106:667–72. [DOI] [PubMed] [Google Scholar]
- 24. Nevins T, Su Y, Doucette S, Kanigsberg N. Incidence of cutaneous malignant melanoma in the Ottawa region: 1996 to 2006. J Cutan Med Surg. 2008;12:276–81. [DOI] [PubMed] [Google Scholar]
- 25. Iannacone MR, Youlden DR, Baade PD, Aitken JF, Green AC. Melanoma incidence trends and survival in adolescents and young adults in Queensland, Australia. Int J Cancer. 2015;136:603–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Mocellin S, Nitti D. Cutaneous melanoma in situ: translational evidence from a large population‐based study. Oncologist. 2011;16:896–903. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Youlden DR, Youl PH, Soyer HP, Aitken JF, Baade PD. Distribution of subsequent primary invasive melanomas following a first primary invasive or in situ melanoma Queensland, Australia, 1982–2010. JAMA Dermatol. 2014;150:526–34. [DOI] [PubMed] [Google Scholar]
- 28. Helvind NM, Hölmich LR, Smith S, Glud M, Andersen KK, Dalton SO, et al. Incidence of in situ and invasive melanoma in Denmark from 1985 through 2012: a national database study of 24 059 melanoma cases. JAMA Dermatol. 2015;151:1087–95. [DOI] [PubMed] [Google Scholar]
- 29. Crocetti E, Caldarella A, Chiarugi A, Nardini P, Zappa M. Does in situ melanoma really come before invasive melanoma? Descriptive epidemiology questions this relationship. Tumori. 2011;97:257. [DOI] [PubMed] [Google Scholar]
- 30. McCaul KA, Fritschi L, Baade P, Coory M. The incidence of second primary invasive melanoma in Queensland, 1982‐2003. Cancer Causes Control. 2008;19:451–8. [DOI] [PubMed] [Google Scholar]
- 31. van der Leest RJ, Liu L, Coebergh JW, Neumann HA, Mooi WJ, Nijsten T, et al. Risk of second primary in situ and invasive melanoma in a Dutch population‐based cohort: 1989‐2008. Br J Dermatol. 2012;167:1321–30. [DOI] [PubMed] [Google Scholar]
- 32. Pomerantz H, Huang D, Weinstock MA. Risk of subsequent melanoma after melanoma in situ and invasive melanoma: a population‐based study from 1973 to 2011. J Am Acad Dermatol. 2015;72:794–800. [DOI] [PubMed] [Google Scholar]
- 33. Claeson M, Holmström P, Hallberg S, Gillstedt M, Gonzalez H, Wennberg AM, et al. Multiple primary melanomas: a common occurrence in Western Sweden. Acta Derm Venereol. 2017;97:715–9. [DOI] [PubMed] [Google Scholar]
- 34. Schuurman MS, de Waal AC, Thijs EJM, van Rossum MM, Kiemeney LALM, Aben KKH. Risk factors for second primary melanoma among Dutch patients with melanoma. Br J Dermatol. 2017;176:971–8. [DOI] [PubMed] [Google Scholar]
- 35. Ministry of Health . New Zealand Cancer Registry (NZCR). 2020. [cited 2020 Jan 13]. Available from: https://www.health.govt.nz/nz‐health‐statistics/national‐collections‐and‐surveys/collections/new‐zealand‐cancer‐registry‐nzcr [Google Scholar]
- 36. Youlden DR, Khosrotehrani K, Green AC, Soyer HP, Kimlin MG, Youl PH, et al. Diagnosis of an additional in situ melanoma does not influence survival for patients with a single invasive melanoma: a registry‐based follow‐up study. Australas J Dermatol. 2016;57:57–60. [DOI] [PubMed] [Google Scholar]
- 37. Bradburn MJ, Clark TG, Love SB, Altman DG. Survival analysis part II: multivariate data analysis ‐‐ an introduction to concepts and methods. Br J Cancer. 2003;89:431–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Schoenfeld D. Partial residuals for the proportional hazards regression model. Biometrika. 1982;69:239. [Google Scholar]
- 39. van der Leest RJT, Zoutendijk J, Nijsten T, Mooi WJ, van der Rhee JI, de Vries E, et al. Increasing time trends of thin melanomas in the Netherlands: what are the explanations of recent accelerations? Eur J Cancer. 2015;51:2833–41. [DOI] [PubMed] [Google Scholar]
- 40. Cust AE, Badcock C, Smith J, Thomas NE, Haydu LE, Armstrong BK, et al. A risk prediction model for the development of subsequent primary melanoma in a population‐based cohort. Br J Dermatol. 2020;182:1148–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Australian Cancer Network Melanoma Guidelines Revision Working Party . Clinical practice guidelines for the management of melanoma in Australia and New Zealand. Wellington: The Cancer Council Australia and Australian Cancer Network, Sydney and New Zealand Guidelines Group; 2008. [Google Scholar]
- 42. Dummer R, Siano M, Hunger RE, Lindenblatt N, Braun R, Michielin O, et al. The updated Swiss Guidelines 2016 for the treatment and follow‐up of cutaneous melanoma. Swiss Med Wkly. 2016;146:w14279. [DOI] [PubMed] [Google Scholar]
- 43. MelNet . Quality statements to guide melanoma diagnosis and treatment in New Zealand: draft for consultation. Christchurch: Melanoma Network of New Zealand (MelNet); 2021. [Google Scholar]
- 44. Cromwell KD, Ross MI, Xing Y, Gershenwald JE, Royal RE, Lucci A, et al. Variability in melanoma post‐treatment surveillance practices by country and physician specialty: a systematic review. Melanoma Res. 2012;22:376–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Garbe C, Hauschild A, Volkenandt M, Schadendorf D, Stolz W, Reinhold U, et al. Evidence and interdisciplinary consense‐based German guidelines: diagnosis and surveillance of melanoma. Melanoma Res. 2007;17:393–9. [DOI] [PubMed] [Google Scholar]
- 46. Marsden JR, Newton‐Bishop JA, Burrows L, Cook M, Corrie PG, Cox NH, et al. Revised U.K. guidelines for the management of cutaneous melanoma 2010. Br J Dermatol. 2010;163:238–56. [DOI] [PubMed] [Google Scholar]
- 47. Ackermann DM, Dieng M, Medcalf E, Jenkins MC, van Kemenade CH, Janda M, et al. Assessing the potential for patient‐led surveillance after treatment of localized melanoma (MEL‐SELF): a pilot randomized clinical trial. JAMA Dermatol. 2021;158:33–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Ackermann DM, Smit AK, Janda M, van Kemenade CH, Dieng M, Morton RL, et al. Can patient‐led surveillance detect subsequent new primary or recurrent melanomas and reduce the need for routinely scheduled follow‐up? A protocol for the MEL‐SELF randomised controlled trial. Trials. 2021;22:324. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49. Stats NZ. Provisional migration estimates by age‐sex now available. 2022. [cited 2022 Mar 18]. Available from: https://www.stats.govt.nz/news/provisional‐migration‐estimates‐by‐age‐sex‐now‐available#fig‐2 [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting Information S1
Data Availability Statement
The data for this study were provided by the New Zealand Ministry of Health, and may be available to other researchers who meet data access requirements. Please contact data_enquiries@moh.govt.nz for further details on eligibility and data provision.